When deploying test execution tools, one major consideration is that recorded manual tests may become unstable in use. This instability can occur due to changes in the application's user interface or other environmental factors, making the recorded scripts unreliable or requiring frequent updates. Data-driven and keyword-driven testing are common methods used with test execution tools, and expected results still need to be defined since tools generally do not generate them automatically.